Constitutional AI Quiz

Answers and explanations appear as you take the quiz.

1. What is the 'constitution' in Constitutional AI?
   • A legal contract between the user and the company
   • The model's neural network architecture
   • An explicit written set of principles the model uses to critique and guide its own responses
   • A list of banned vocabulary words
2. What annotation burden does the original Constitutional AI study aim to reduce?
   • Every possible source of human judgment or training data
   • All model memory during inference
   • The need to evaluate any model behavior
   • Reliance on human labels identifying harmful outputs
3. In the supervised stage of Constitutional AI, what does the model do to its own output?
   • Deletes it without explanation
   • Critiques it against a principle and then revises it to be better
   • Translates it into a different language
   • Encrypts it for safety
4. What does RLAIF stand for in the reinforcement-learning stage?
   • Reinforcement Learning from AI Feedback
   • Recursive Learning Among Internal Functions
   • Reduced Latency AI Framework
   • Random Label Assignment In Fine-tuning
5. What transparency benefit can written constitutional principles provide?
   • A complete display of every learned internal preference
   • Proof that every future response follows all principles
   • Automatic elimination of the preference model
   • An inspectable statement of intended guidance used in the process
6. Which broader research idea does using AI-generated supervision support?
   • Eliminating every evaluation by definition
   • Exploring how AI can help supervise other AI systems
   • Guaranteeing zero cost for all future training
   • Automatically increasing every model’s vocabulary
7. Does editing a constitution document alone retroactively change the weights of an already saved trained model?
   • No; a training or runtime process must use the changed guidance
   • Yes; all saved checkpoints instantly rewrite themselves
   • Yes; written words directly become new neural weights without execution
   • No model process can ever consume written principles
8. Which human comparison data did the original Constitutional AI RL experiments retain alongside AI harmlessness preferences?
   • Only human GPU-temperature measurements
   • No human comparison data of any kind
   • Human helpfulness comparisons
   • A mandatory human label for every harmlessness pair